AKUSTIK MODЕLLASH VA OVOZ XUSUSIYATLARINI CHIQARISH.

Mamatov N. S., Nuritdinov N. D. & Madaminjonov A. D.


Annotatsiya

Ushbu maqolada akustik modellashtirish va ovoz xususiyatlarini chiqarishning asosiy tamoyillari tahlil qilinadi. Avvalo, nutq signalining akustik xususiyatlar, akustik xususiyatlarni ifodalash usullari ko‘rib chiqildi. Shuningdek, ovoz signalidan muhim xususiyatlarni — Mel-chastotali kepstral koeffitsiyentlar (MFCC), spektral og‘irlik markazi (spectral centroid), spektral kenglik (bandwidth), spektral tekislik (flatness) va boshqa parametrlarni aniqlash metodologiyasi bayon etiladi. Ushbu parametrlar nutqni matnga o‘tkazish, matndan nutqga aylantirish va ovozli identifikatsiya tizimlarida muhim rol o‘ynaydi. Maqolada, shuningdek, Python dasturlash tili va librosa, numpy, scipy kutubxonalari yordamida ovoz xususiyatlarini avtomatik ajratish algoritmlari amaliy misollar bilan ko‘rsatib berildi. Tadqiqot natijalari ovoz asosidagi intellektual tizimlarni ishlab chiqishda qo‘llanilishi mumkin.


Adabiyotlar

  1. Gulati, A., Qin, J., Chiu, C.-C., et al. (2020). Conformer: Convolution-augmented Transformer for Speech Recognition. https://arxiv.org/abs/2005.08100
  2. Hsu, W.-N., et al. (2021). Hubert: Self-supervised Speech Representation Learning by Masked Prediction of Hidden Units. https://arxiv.org/abs/2106.07447
  3. Radford, A., et al. (2023). Whisper: Robust Speech Recognition via Large-Scale Weak Supervision. https://arxiv.org/abs/2212.04356
  4. Baevski, A., Zhou, H., Mohamed, A., & Auli, M. (2020). wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations. https://arxiv.org/abs/2006.11477
  5. Babu, A., et al. (2022). XLS-R: Self-Supervised Cross-Lingual Speech Representation Learning at Scale. https://arxiv.org/abs/2111.09296
  6. Zhao, J., Wang, Y., & Lee, C. H. (2021). Robust Feature Extraction for Speech Recognition Using Deep Learning. IEEE Transactions on Audio, Speech, and Language Processing. https://doi.org/10.1109/TASLP.2021.3059876
  7. Tao, J., Zhang, R., & Xu, B. (2022). Speech Enhancement via Mel Spectrogram Learning with CNNs. Computer Speech & Language, Elsevier. https://doi.org/10.1016/j.csl.2022.101282
  8. Lai, W., Yu, Z., & Liu, X. (2023). Spectral Features for Emotion Recognition Using Neural Networks. IEEE Access, 11, 23456–23465. https://doi.org/10.1109/ACCESS.2023.3241234
  9. Park, D. S., Chan, W., Zhang, Y., et al. (2020). SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition. Interspeech 2020. https://www.isca-speech.org/archive/Interspeech_2020/abstracts/3096.html
  10. https://www.geeksforgeeks.org/nlp/mel-frequency-cepstral-coefficients-mfcc-for-speech-recognition/#mfcc-for-speech-recognition-faqs
  11. https://en.wikipedia.org/wiki/Short-time_Fourier_transform